Abstract

This paper asks whether machine learning can forecast euro area sovereign bond spreads and whether the resulting forecasts can track financial fragmentation. Using a new high-dimensional monthly dataset of 4,948 macro-financial series for ten euro area countries from December 2008 to February 2025, we run a horse race among thirteen machine-learning models and two simple benchmarks, an AR(1) process and a random walk. XGBoost is the strongest machine-learning model and is never significantly outperformed by the other learners. It is not, however, the most accurate one-month-ahead forecaster: under strict out-of-sample re-estimation the AR(1) and the random walk attain lower point-forecast errors in every country. The value of the machine-learning approach lies elsewhere. SHAP decompositions recover the macro-financial drivers of the predicted spreads. The forecasts form the basis of a fragmentation indicator built by clustering predicted spread paths. The indicator reproduces the core-periphery divide in the windows running to 2022. French spreads then decouple from the core after 2023. By 2024 to 2025 France and Belgium form a distinct cluster, a new source of fragmentation risk with direct implications for the transmission of a single monetary policy.

Keywords: Euro area sovereign spreads; Financial fragmentation; Machine learning; Forecasting; XGBoost; SHAP

Abstract

This paper studies the financial fragmentation risk in the Euro area sovereign bond market using a combination of machine learning forecasting and network analysis. By applying the XGBoost model to predict sovereign bond yield spreads and constructing correlation-based networks, our analysis identifies structural patterns of financial fragmentation across different forecast horizons. The results indicate that core Euro area economies -such as Austria, Finland and the Netherlands- exhibit strong co-movements in their sovereign spreads, while peripheral countries -including Greece, Ireland and, at longer horizons, France- display weaker correlations. Network analysis highlights key intermediaries in sovereign risk transmission, with Italy and Greece showing strong linkages, while Ireland and Belgium appear to be disconnected. Additionally, we designed a fragmentation indicator, based on betweenness and closeness centrality, that quantifies the degree of financial segmentation over time. Despite policy efforts to enhance financial integration, our findings suggest that fragmentation remains a persistent feature of the Euro area sovereign bond market.

Keywords: Machine Learning, Financial Fragmentation, Network Centrality, Sovereign Spreads, Euro Area

Abstract

The euro area shares a common monetary policy but features persistent cross-country heterogeneity in banking-sector structure and credit conditions. This paper examines whether such heterogeneity generates real fragmentation by altering the transmission of ECB monetary policy shocks to private investment. We combine identified ECB monetary policy shocks (covering conventional and unconventional measures) with quarterly country-sector investment data and estimate impulse responses using panel local projections. To capture structure-dependent transmission, we interact monetary policy shocks with pre-determined measures of credit structure, including bank concentration, non-performing loan ratios, capital adequacy, funding composition, and survey-based credit standards. We then propose an operational measure of investment fragmentation defined as the cross-sectional dispersion of cumulative investment responses to a common shock, and construct a time-varying fragmentation index that can be decomposed into a component explained by credit-structure differences and a residual component attributable to fundamentals and other frictions. The findings aim to contribute to policy discussions on the optimal design of ECB interventions and financial integration strategies, addressing challenges related to economic convergence and monetary policy efficiency within the Euro area.

Keywords: Monetary Policy, Credit Structure, Financial Fragmentation, Euro Area, Bank Lending